Understanding behaviour of District heating systems Integrating Distributed sources

Future district heating and cooling systems will be based on completely renewable energies from solar, waste heat, and geothermal energy. This will imply that many distributed systems have to be available to deliver heat to the central system. A system like this is called distributed since there is no common central neither a common owner. In this new situation, a building will be able to be a heat user and supplier at the same time or so-called prosumer. Current theoretical background and methods for the energy system analysis are based on the centralized assumption. The current theories do not consider bidirectional flows and integrated management between centralized and distributed supply and demand. In addition, in a water-based energy system, there is a natural monopoly of the main system. Therefore, it is even more difficult for these new distributed sources to be integrated to the system. Therefore, the aim of this project was to develop models to explain interfaces between the energy supply and demand in a district heating integrating distributed sources. The term interfaces may be explained via hard and soft issues. The hard issues cover the following topics: district heat network, requirements for consumer substations and buildings, and connection principles for decentralized sources. The soft issues cover the following topics: technical and economical modelling of the distribution system, optimization between demand and supply, innovative and integrated control concepts and energy management, transition of the existing district heating grid to the low temperature grid, and new pricing and business models. Within UnDID project, three methods to model the district heating grid were developed: 1) linear approach, 2) approach based on the momentum and energy equations, and 3) dynamic models in Modelica. These methods were implemented on a synthetic district heating grid supplying low energy buildings and to model district heating grid at the Gløshaugen campus at NTNU in Trondheim. Analysis of the distributed heat sources focused on better utilization of locally available heat sources by implementing short- and long-term thermal storages. The results of the district heating analysis indicated that decreasing the return temperature is a key issue on the way to include more renewables. The results on the temperature control strategies for building heating system indicate that there are lots of potentials in better control of the building supply temperature. Within this project, model predictive control (MPC) was considered for the system control on building level and on the district heating level. MPC is an advanced control method used to control a process while satisfying a set of constraints. Within the project comprehensive analyses have been performed to investigate how to utilize the waste heat from distributed sources. The analysis of the pressure and temperature level showed that the connection type of the distributed sources may influence a lot the amount of harvested waste heat. For example, connecting waste heat from a data center to the return line would induce higher heat losses and the higher return temperature. In our analysis, we came that connecting the waste heat to the supply line is a better solution, because it helped to avoid pressure cones. Further, the dynamic analysis of the district heating system showed that short- and long-term storage may help to better utilize the waste heat and decrease unnecessary high peaks. However, when considering the investment, the best way to utilize the waste heat is with the short-term storages. Within this project a case of a small district heating grid at the NTNU university campus was analyzed. Data center as a heat prosumer was observed because this building is using and giving heat at the same time. As explained above different hydraulic and thermal analyses were performed. The results showed that a data center may be a reliable heat deliverer. Utilization of a short-term water storage tank may help to improve both economic and energy performance of such systems. However, price models in district heating will always influence the decision about the size of the short-term water storage. Finally, regardless of the prices, it is worth to utilize the short-term water storage on daily basis. Finally, MPCs were developed to optimize building control and for the entire district heating grid at the NTNU university campus. Results showed that introducing the error model for the MPC scheme was able to address the weather forecast uncertainty and hence achieve almost the full theoretical potential of the MPC in terms of heating cost-saving and indoor temperature control. Finally, when MPC was implemented for the entire district heating grid, the MPC scheme made an optimized trade-off between the heat and electricity use to achieve the best economic performance of the heat prosumer.

Internationally, we have joined one project under International Energy Agency within the District heating and cooling program, TS2 - Implementation of Low-Temperature District Heating Systems. Throughout the project, we have established a strong collaboration with the research institute Lawrence Berkeley National Laboratory, USA, and Professor Gongsheng Huang from City University of Hong Kong, Hong Kong. These resulted in many high-quality publications. Within this project, we participated at one industrial workshop to present our results. Finally, Professor Natasa Nord was invited to the keynote speaker at two conferences, Applied Energy in Sweden (hold on-line) and Thermal power plants in Serbia (hold on-line). Based on our work in the UnDID project, Professor Natasa Nord is involved in an internally funded project by Faculty of Engineering at NTNU with the focus are sustainability, Optimal Utilization of Resources towards Neutral Climate Built Environments in Europe by 2030-2050 (LIFELINE-2050). The focus of this project is to achieve resource efficient and climate neutral built environment with the NTNU campus as a case. Within this project, Professor Natasa Nord will be a work package leader and a PhD supervisor.

Use of renewable energies and waste energy is highly necessary and required by national and international regulations. Future district heating and cooling systems will be based on completely renewable energies from solar, waste heat, and geothermal energy. This will imply that many distributed systems have to be available to deliver their heat to the central system. In this new situation, a building will be able to be user and supplier at the same time. Current theoretical background and methods for the energy system analysis are based on the centralized assumption that there are lots of demands but only one supply to fulfill all the demands. The current theories do not consider bidirectional flows and integrated management between centralized and distributed supply and demand. In addition, in a water-based energy system, there is a natural monopoly of the main system. Therefore, it is even more difficult for these new distributed sources to be integrated to the system. One of the main reasons why this research topic is still new and not well developed is that income and cost models for sharing the benefit of energy efficiency at supply and demand side are not common and do not support fully environmental friendliness that is an ultimate goal. Therefore, the aim of this project is to develop models to explain interfaces between the energy supply and demand in a district heating integrating distributed sources. To develop fundamental models for this problem, the scientific methods combining physical models, modeling, building energy simulations, measurements, statistical methods, process simulation, optimization, and control theory will be implemented. Current design, control, energy planning methods, and fundamental background have not been treated this new situation with the distributed heat sources. Hence consultants, building operators, property owners, district heating companies, and energy planners have a huge knowledge need.